"Sometimes the probabilities are very close to certainties, but they're never really certainties"
About this Quote
Certainty is a seductive illusion; the world we inhabit, especially as science studies it, is probabilistic. Murray Gell-Mann, who helped uncover quarks and later explored complexity at the Santa Fe Institute, knew that even the best-supported statements about nature come with error bars. Quantum mechanics makes randomness foundational, but the point is broader: models abstract a messy reality, data are finite and noisy, and every inference carries assumptions that may fail at the margins.
Calling something 99.9999 percent likely is not the same as declaring it beyond doubt. Particle physicists famously wait for five-sigma significance before announcing a discovery, yet five sigma is still a probability, not a proof. Weather forecasts, medical diagnoses, and election models all speak in odds because they are negotiating between incomplete information and the need to act. Good practice builds in margins, redundancies, and continual updating. That is not weakness; it is rigor and humility.
The human mind craves certainty and is prone to overconfidence. Treating very high probabilities as guarantees fuels complacency and brittle systems. Low-probability events, from financial crashes to rare side effects, do happen, and when the stakes are large, the tail matters. Conversely, refusing to act until there is absolute certainty paralyzes decisions in domains where waiting increases risk. Policy on climate, public health, and infrastructure must operate under uncertainty, weighing consequences along with likelihoods.
There is a philosophical edge here. Mathematics can deliver certainties within axioms, but empirical knowledge remains provisional. The scientific attitude accepts that every claim is open to revision in light of new evidence, and that is why science advances. Gell-Mann’s reminder encourages clarity about confidence without pretending to omniscience. It asks us to quantify uncertainty, communicate it honestly, and make choices that are robust to being wrong, because reality gives us probabilities that may be very close to certainties, but never the thing itself.
Calling something 99.9999 percent likely is not the same as declaring it beyond doubt. Particle physicists famously wait for five-sigma significance before announcing a discovery, yet five sigma is still a probability, not a proof. Weather forecasts, medical diagnoses, and election models all speak in odds because they are negotiating between incomplete information and the need to act. Good practice builds in margins, redundancies, and continual updating. That is not weakness; it is rigor and humility.
The human mind craves certainty and is prone to overconfidence. Treating very high probabilities as guarantees fuels complacency and brittle systems. Low-probability events, from financial crashes to rare side effects, do happen, and when the stakes are large, the tail matters. Conversely, refusing to act until there is absolute certainty paralyzes decisions in domains where waiting increases risk. Policy on climate, public health, and infrastructure must operate under uncertainty, weighing consequences along with likelihoods.
There is a philosophical edge here. Mathematics can deliver certainties within axioms, but empirical knowledge remains provisional. The scientific attitude accepts that every claim is open to revision in light of new evidence, and that is why science advances. Gell-Mann’s reminder encourages clarity about confidence without pretending to omniscience. It asks us to quantify uncertainty, communicate it honestly, and make choices that are robust to being wrong, because reality gives us probabilities that may be very close to certainties, but never the thing itself.
Quote Details
| Topic | Truth |
|---|
More Quotes by Murray
Add to List
